This application claims the priority of Korean Patent Application No. 2004-48041, filed on Jun. 25, 2004, in the Korean Intellectual Property Office, the disclosure of which is incorporated herein in its entirety by reference.
1. Field of the Invention
The present invention relates to a memory device, and more particularly, to a wake-up method of and a wake-up circuit for converting a sleep mode to an active mode in a memory device.
2. Description of the Related Art
In a semiconductor memory device, it is very important to prevent malfunction that is generated as a result of noise in the power supply. There are several sources of noise when a semiconductor memory device operates. A peak current, which is generated when discharged parts of a circuit are simultaneously precharged, is a typical source if such noise.
In particular, in a memory cell array, there is a high possibility that large peak currents will be generated in a bit line pair when reading data of a bit cell.
One of the primary considerations regarding the type of memory that is commonly used in portable telephone systems is amount of loss due to leakage current. To this end, several architectures have been proposed, and an architecture in which leakage current is reduced by placing the memory device in a sleep mode that deactivates the application of the power supply voltage to the memory device is widely used. However, a large amount of peak current is generated during a wake-up operation in which the memory device transitions from a sleep mode to an active mode. The peak current causes an IR drop due to the resistance of the power line, and the power supply voltage supplied to the memory cell therefore drops. In a worst-case scenario, the power supply voltage can drop below a retention voltage of the memory cell, which can therefore result in data loss of the memory cell.
FIGS. 1 and 2 are circuit diagrams of wake-up circuits of a conventional SRAM (static random access memory) memory device.
In a power supply voltage power-off structure, which is used to implement a low-leakage current SRAM, all bit lines are precharged from a ground voltage to the power supply voltage using a wake-up operation when switching the SRAM from a sleep mode to a stand-by mode during the wake-up operation. During this operation, a peak current is generated, in turn causing a voltage drop.
In the wake-up circuit as shown in FIG. 1, all precharge circuits 10 included in a column simultaneously operate during the wake-up operation. However, when a control signal SC is applied as shown in FIG. 1, if all of the precharge circuits 10 connected to the control signal line 12 simultaneously start operating, all of the bit line pairs respectively connected to the precharge circuits 10 are concurrently precharged to the power supply voltage, and thereby the power supply voltage supplied to the memory drops. That is, data of the SRAM bit cell is threatened by the IR drop occurring due to the reduction of the power supply voltage.
The wake-up circuit shown in FIG. 2 distributes the peak currents to resolve the problems that can occur in the wake-up circuit illustrated in FIG. 1. Referring to FIG. 2, a plurality of bit line pairs which are connected to the memory cell are divided into a plurality of blocks 2_1, 2_m, . . . , 2_n. In addition, each of the plurality of divided blocks includes a plurality of inverter chains 26, 28 . . . to distribute the peak currents.
A control signal SC is input via a wake-up control line 24 and precharges the bit line pairs included in the first block 2_1. Specifically, the control signal SC is input to a precharge circuit 20 included in the first block 2_1 to precharge the bit lines connected to the precharge circuit 20 to the power supply voltage VDD. The control signal SC applied to the first block 2_1 is delayed by inverter chain 26 and a delayed control signal SC1 is output by the inverter chain 26 and input to a precharge circuit 22 included in the second block 2_2 to precharge the bit lines connected to the precharge circuit 22 to the power supply voltage VDD. The delayed control signal SC1 is then delayed by inverter chain 28, and a second delayed control signal SC2 output by the inverter chain 28 precharges bit line pairs of the subsequent block.
The inverter chains 26, 28, allow the wake-up operation to commence in a subsequent block without determining whether the wake-up operation has been completed in a previous block. Therefore, the peak currents may not be properly distributed. Also, when the delay time of the inverter chains 26, 28, are increased to preserve data of the memory cell, timing loss can result from an increased wake-up timing margin.
FIG. 3 is a timing diagram of the control signals SC, SC1 and SC2 of the wake-up circuit shown in FIG. 2. The control signal SC is output from a control signal outputting unit (not shown), and the control signal SC1 output from the inverter chain 26 is delayed by a delay time d after the control signal SC1 is output.
As shown in FIG. 3, each of the inverter chains delays the input control signal for the delay time d regardless of whether the associated bit lines wake-up.
FIG. 4 is a graph illustrating the generated power supply voltage when the wake-up circuit shown in FIG. 2 is in a wake-up mode. Referring to FIG. 4, when using the inverter chains 26, 28, having the fixed delay times d illustrated in FIG. 2, since bit lines of subsequent blocks start waking up before bit lines of previous blocks have sufficiently awoken, an IR drop occurs and, as shown in FIG. 4, the power supply voltage (VDD) drops to a low voltage. Alternatively, when the delay time in the inverter chain is increased by increasing the number of inverters, the overall time required for the wake-up operation is increased.